Microsoft restores services after global outage hit 70-plus countries

Microsoft declared its services fully restored on Saturday morning, closing the book on one of the worst IT failures in years — an outage that disrupted governments, courts, hospitals, airlines and banks across more than 70 countries and territories.

The company confirmed the fix on its Microsoft 365 status page. "We have completed our efforts to address the underlying issue of the incident and restored service to the remaining impacted customers and regions," Microsoft wrote. The firm pointed administrators to follow-up information under reference MO1550821 in the Microsoft 365 admin center.

The scale was immediate and measurable. Downdetector, an online outage tracker, said it received reports from 74 countries and territories within a single hour of the first problems emerging. Users lost access to core Microsoft 365 services including Outlook, Teams and the AI assistant Copilot.

More than 430 million people use Microsoft 365 apps, according to figures Microsoft publishes on its website. In January, the company separately said it had more than 450 million paid commercial subscriptions to its 365 services. The total number of people affected by this week's outage was not clear, but the user base alone frames the stakes.

When did the outage start?

Problems first appeared at about midday on Thursday in the US, which was 6pm in the UK. Governments in Japan, Singapore and Australia reported that the hours-long episode affected some of their operations. Banks, airlines, businesses and healthcare services reported disruption in many other places.

In the UK, the outage hit vital public services hard. Hospital staff resorted to paper records. Courtrooms scrambled to locate fax machines — a detail that captured how far some institutions have to reach back when modern cloud tools go dark.

What caused it?

Microsoft apologised on Friday and moved to head off speculation about the cause. The company said the outage was "not related to cyber" or artificial intelligence agents. Instead, it traced the failure to internal work: the incident "was introduced by an issue during an earlier network capacity upgrade that surfaced" after a recent network change.

That explanation matters for the broader debate. The outage was not a hack, and it was not the AI agents Microsoft has woven into its productivity suite. It was, by the company's own account, a routine infrastructure change gone wrong — a mundane failure mode that still produced global consequences.

Why does concentration worry policymakers?

The disruptions offer the latest example of the cascading problems that arise when so many communications and technology platforms depend on the same provider. One faulty network change at a single vendor produced courtroom delays in the UK and government service interruptions on three continents.

In Britain, the outage fuelled concerns about over-reliance on a small number of technology companies. Tim Clement-Jones, a former chair of the House of Lords artificial intelligence select committee, called the widespread outages of vital government services an "I-told-you-so moment" — a pointed marker of Britain's dependence on US tech.

Academics reached the same conclusion from a different angle. Jeremy Singer, a professor at the University of Glasgow's school of computing science, said: "This problem serves to emphasise the dependence we have built on one company's digital infrastructure, and may lead us to question whether we need more resilience in our IT ecosystem."

That question — resilience versus consolidation — is not new, but each major outage sharpens it. When a hospital's fallback is a paper chart and a courtroom's fallback is a fax machine, the redundancy that once existed in distributed IT systems has effectively eroded.

What comes next?

Microsoft has restored service and published an incident reference for administrators. What it has not yet quantified is the total number of users affected across the outage window, which began Thursday at midday US time and stretched through Friday before the Saturday-morning resolution of the remaining impacted regions.

The political pressure, at least in the UK, is unlikely to subside with the status page. Clement-Jones's "I-told-you-so moment" signals that lawmakers who have warned about Big Tech concentration now have a fresh, concrete case to cite. Singer's framing points in the same direction: the episode may push institutions to ask whether their IT ecosystems need built-in redundancy rather than single-provider defaults.

For Microsoft, the restoration closes the operational incident but opens a reputational one. A company with more than 450 million paid commercial subscriptions just demonstrated, to governments and hospitals alike, exactly what a single network change can take down.

---

By Elena Vasquez